Physicochemical and Functional Characterization of the Collagen

Jul 22, 2014 - Marı́a las Torres, Cuautitlán Izcalli, Estado de México C.P. 54740, ... Contreras-Figueroa , GE Cortina-Ramírez , René Fernando A...
0 downloads 0 Views 8MB Size
Article pubs.acs.org/JPCB

Physicochemical and Functional Characterization of the Collagen− Polyvinylpyrrolidone Copolymer Gerardo Leyva-Gómez,† Enrique Lima,‡ Guillermo Krötzsch,§ Rosario Pacheco-Marín,∥ Nayeli Rodríguez-Fuentes,⊥ David Quintanar-Guerrero,# and Edgar Krötzsch*,† †

Laboratory of Connective Tissue, Centro Nacional de Investigación y Atención de Quemados, Instituto Nacional de Rehabilitación, Mexico City, Mexico ‡ Instituto de Investigaciones en Materiales, Universidad Nacional Autónoma de México, Circuito exterior s/n, Cd. Universitaria, Del. Coyoacán, C.P. 04510 México D. F., Mexico § Instituto de Ciencias Fı ́sicas, Universidad Nacional Autónoma de México, Av. Universidad s/n, Col. Chamilpa, C.P. 62210, Cuernavaca City, Mexico ∥ Laboratory of Molecular Biology, Instituto Nacional de Medicina Genómica, Periférico Sur 4124, Torre Zafiro II 5to piso, Col. Ex-Rancho de Anzaldo, Mexico City, Mexico ⊥ Departamento de Materiales Metálicos y Cerámicos, Instituto de Investigaciones en Materiales, Universidad Nacional Autónoma de México, Circuito exterior s/n, Cd. Universitaria, Del. Coyoacán, C.P. 04510, México D. F., Mexico # Laboratorio de Investigación y Posgrado en Tecnologı ́a Farmacéutica, Facultad de Estudios Superiores Cuautitlán, Universidad Nacional Autónoma de México, Av. 1° de Mayo s/n, Col. Sta. Marı ́a las Torres, Cuautitlán Izcalli, Estado de México C.P. 54740, Mexico ABSTRACT: Collagen−polyvinylpyrrolidone (C−PVP) is a copolymer that is generated from the γ irradiation of a mixture of type I collagen and low-molecular-weight PVP. It is characterized by immunomodulatory, fibrolytic, and antifibrotic properties. Here, we used various physicochemical and biological strategies to characterize the structure, biochemical susceptibility, as well as its effects on metabolic activity in fibroblasts. C−PVP contained 16 times more PVP than collagen, but only 55.8% of PVP was bonded. Nevertheless, the remaining PVP exerted strong structural activity due to the existence of weak bonds that provided shielding in the NMR spectra. On SEM and AFM, freeze-dried C− PVP appeared as a film that uniformly covered the collagen fibers. Size analysis revealed the presence of abundant PVP molecules in the solution of the copolymer with a unique dimension related to macromolecular combinations. Calorimetric analysis showed that the copolymer in solution exhibited structural changes at 110 °C, whereas the lyophilized form showed such changes at temperatures below 50 °C. The copolymer presented a rheopectic behavior, with a predominant effect of the collagen. C−PVP had biological effects on the expression of integrin α2 and prolyl-hydroxylase but did not interact with cells through the collagen receptors because it did not inhibit or slow contraction.



INTRODUCTION Collagen−polyvinylpyrrolidone (C−PVP) is a compound derived from the γ irradiation of pepsinized porcine type I collagen fibers and low-molecular-weight (LMW) polyvinylpyrrolidone (PVP). The physicochemical and pharmacological properties of this copolymer are derived from its components. C−PVP exhibits immunomodulatory and tissue-regenerative properties, which have been evidenced in different pathologies in acute and chronic stages in various fields, including orthopedics and trauma,1−4 vascular surgery,5 gynecology and gastroenterology,6−8 as well as dermatology and rheumatology.9−15 However, little is currently known about the structure of this copolymer. Polyacrylamide gel electrophoresis (PAGE) anal© 2014 American Chemical Society

yses have shown that C−PVP has a slight decrease versus collagen in the relative mobilities of the corresponding species to the α1 (I) and α2 (I) collagen chains.16 High-performance liquid chromatography (HPLC) has shown that collagen has two species, whereas the copolymer and PVP have only one (unpublished observations). These findings, along with the observation that the biological activity of C−PVP differs from that of its individual components, suggest that these structural differences could determine the biological activity.17 Received: March 17, 2014 Revised: June 20, 2014 Published: July 22, 2014 9272

dx.doi.org/10.1021/jp502476x | J. Phys. Chem. B 2014, 118, 9272−9283

The Journal of Physical Chemistry B

Article

Table 1. Summary of the Samples and Assays with Which They Were Tested sample

name

polyvinylpyrrolidone K15 irradiated polyvinylpyrrolidone K15 type I collagen

PVP PVPirr collagen

collagen and polyvinylpyrrolidone collagen− polyvinylpyrrolidone

C+PVP C−PVP

collagen− polyvinylpyrrolidone w/ o excess polyvinylpyrrolidone

C−PVP/ −PVP

assaya

characteristic synthetic polymer Co-irradiated synthetic polymer,10 KGy biological polymer

60

physical mixture (w/o irradiation) pharmaceutical grade copolymer (60Co-irradiated, 10KGy) pharmaceutical grade copolymer (60Co-irradiated, 10KGy), membrane filtration 100-kDa molecular cutoff

EC, SEM, AFM, NMR-MAS, DSC, particle size NMR-MAS, DSC, particle size SEM, AFM, NMR-MAS, DSC, digestion by MMP-1, rheology, particle size AFM, NMR-MAS, DSC, rheology, particle size EC, SEM, AFM, NMR-MAS, DSC, digestion by MMP-1, rheology, particle size, biological activity AFM, DSC, rheology, particle size

a

EC = capillary electrophoresis, SEM = scanning electron microscopy, AFM = atomic force microscopy, NMR-MAS = nuclear magnetic resonance with magic angle spinning, DSC = differential scanning calorimetry, MMP-1 = matrix metalloproteinase 1.

phoretic separation of the samples, a negative capillary of 60 cm (50.3 cm to the detection window) × 50 μm ID was used. Detection was performed at 200 nm. Before each analysis, the capillary was washed for 5 min with running buffer (100 mM phosphate, pH 9.26). Before loading onto the capillary, the sample was diluted with ultrapure water (18 MΩ), injected hydrodynamically (5 s, 0.5 psi), and separated in running buffer at 30 °C and 10 kV. Morphological Analysis. Micrographs were recorded using a Leica Stereoscan 440 scanning electron microscope (SEM) [Leica Microsystems, Bensheim, Germany]. Freeze-dried samples were fixed with 2.5% glutaraldehyde in phosphate buffer (PBS, pH 7.4, 0.1 M) for 1 h and washed with PBS and distilled water to prevent crystal formation. Subsequently, aluminum samples with carbon ribbon lacking a conductive material coating (gold, coal) were placed on a microscope slide. The C−PVP copolymer was also analyzed by atomic force microscopy (AFM) on a JSPM 4210 microscope (JEOL, Tokyo, Japan); Mikromasch silicon cantilevers (NSC14) were used in measurements. A volume of 50 μL of the samples was diluted 1:100 and placed on a coverslip, and the samples were dried at room temperature with a relative humidity of 55%. After 7 days and with a final thickness of 5 μm approximately, the AFM analyses were performed. Every sample was placed on a copper bonding surface and analyzed in the contact mode to measure the topography of the sample in a scanned area close to 5 × 5 μm. The vibration frequency was in the range of 350 kHz. Images were acquired in the amplitude−height mode at room temperature.19 Structure. Liquid 1H nuclear magnetic resonance (NMR) spectra were obtained in solutions diluted in D2O using a Bruker Avance 400 spectrometer (Bruker, Billerica, Massachusetts, U.S.A.) at a frequency of 400.17 MHz. Samples were lyophilized prior to acquisition of 13C NMR spectra with magic angle spinning (MAS NMR) in the solid state. Spectra were acquired on a Bruker Avance II 300 spectrometer (Bruker, Billerica, Massachusetts, U.S.A.) under cross-polarized conditions at a resonance frequency of 75.4 MHz and MAS of 5 kHz, using contact pulses of 100 μs and 1, 3, and 5 ms.20 The average particle size was determined by dynamic light scattering (DLS) on a Zetasizer (ZEN 3600 Nano - ZS, Malvern Instruments, Massachusetts, U.S.A.). The wavelength of the laser (He/Ne, 10 mW) was 633 nm. Measurements were obtained at an angle of 173° for 60 s at 25 °C. The sample concentration was adjusted to 830 μg/mL with a 5 mM acetic acid solution in all cases. Measurements were performed in quadruplicate. Thermal Properties. Differential scanning calorimetry (DSC) was performed on a DSC Q10 (TA Instruments,

The use of type I collagen and PVP as copolymer components is based primarily on their low immunogenicity and tolerance, respectively. Moreover, in vitro and in vivo assays have shown that C−PVP does not stimulate the proliferation of lymphocytes or cause DNA damage, nor does it induce the production of antibodies against porcine type I collagen or C−PVP. No evidence of renal or hepatic failure has been seen in patients receiving weekly doses of 0.2−1.0 mL of C−PVP for 1 year, either intradermally or intramuscularly.18 On the basis of the above evidence, we sought to characterize the structure of the C−PVP copolymer through its physicochemical properties, biological susceptibility, and possible interaction with collagen-specific cell−surface receptors.



MATERIALS AND METHODS Materials. Pepsinized porcine type I collagen (DSM Branch Pentapharm, Aesch, Switzerland) was dialyzed against 5 mM acetic acid (pH 4.3). PVP K-15 (8000 Da; ISP Investments, Inc., Wilmington, DE, U.S.A.) was dissolved in 5 mM acetic acid, and for some specified assays, it was irradiated at 10 kGy using a 60Co pump (Instituto Nacional de Investigaciones Nucleares, Mexico City, Mexico). Both solutions were diluted in proportion to their content in C−PVP copolymer (Fibroquel, Aspid, SA de CV, Mexico City, Mexico), 8.33 and 133 mg/mL, respectively. To assess whether PVP was affected by the irradiation used on the C−PVP copolymer, a fraction of the PVP solution was irradiated under the same conditions as the copolymer. To assess the proportion of PVP integrated into the copolymer, C−PVP was subjected to molecular filtration (Microcon 100 kDa, Millipore, Massachusetts, U.S.A.). For assays requiring materials in the solid state, samples were lyophilized (Freezone 6, Labconco, Missouri, U.S.A.). Table 1 summarizes the samples and assays with which they were tested. Composition. Chromatographic Characterization. HPLC assays were performed in isopropanol−water−trifluoroacetic acid (80%: 19.99%: 0.01% v/v) as the mobile phase through a reverse-phase column (Chrompack P 300 RP, Chrompack, California, U.S.A.; 8 μm, 300 Å, and 150 × 4.6 mm ID). The analysis was run on a Varian chromatograph (Varian, California, U.S.A.) equipped with a Varian INERT 9012 pump and a Varian 9050 UV detector at 243 nm. The equipment was controlled with the Star Chromatography software (version 4.51). The injection volume was 60 μL, and the flow rate was 0.6 mL/min. Electrophoretic Characterization. Capillary electrophoresis analysis was carried out on a ProteomeLab PA 800 (Beckman, California, U.S.A.) using the 32 Karat software. For electro9273

dx.doi.org/10.1021/jp502476x | J. Phys. Chem. B 2014, 118, 9272−9283

The Journal of Physical Chemistry B

Article

A fast Fourier transform (FFT)-based algorithm was used. The representation of the algorithm was previously described by Shin and Jeon24 in the Wahl and Bolton text25

Delaware, U.S.A.) using airtight aluminum sample holders. The heating rate was 10 °C/min for samples in solution and 1 °C/ min for lyophilized samples, with a range of 0−200 °C and a nitrogen flow of 50 mL/min. Each determination was performed in triplicate with only one scan. Enthalpy changes were determined from the experimental data.21 Susceptibility to Matrix Metalloproteinase (MMP)-1. Fibrillar collagens are biologically digested in two steps. First, the quaternary structure is denatured by interstitial collagenases, followed by digestion through gelatinases. Both enzyme groups belong to the family of MMPs.22 An assay of enzymatic digestion of collagen and C−PVP by MMP-1 was conducted. Then, 400 μg of the collagen sample or an equivalent amount of C−PVP was incubated in the presence of 10 ng of MMP-1 (Sigma, Missouri, U.S.A.) in a tris-HCl buffer (pH 7.4) with CaCl2, at 35 °C. Samples were incubated for different time periods (1−60 min). The reaction was interrupted by adding 20 μL of 0.1 M ethylenediaminetetraacetic acid (EDTA) to the reaction medium. Samples were centrifuged at 9000g at 4 °C for 10 min. Supernatants were tested in acrylamide gels at 10%. High-molecular-weight markers were run in the same gel to determine the size of the digested fragments. Rheology and Frequency Spectra Analysis. Viscosity measurements were performed at 25 °C in 200 μL of sample, with the No. 1 needle of a Brookfield CAP 2000 viscometer (Brookfield, Massachusetts, U.S.A.). Each sample was analyzed in triplicate. The equipment was calibrated with standard solution No. CAP3L (Brookfield, Lot. No. 00601). The shear rate experiment was conducted from 1600 to 7999.98 s−1 (120−600 rpm) and vice versa at intervals of 1600 s−1. All solutions were presheared at 1600 s−1 (120 rpm) for 5 min after loading into the concentric conical geometry. Each subsequent recording was taken after spinning for 20 s at each shear rate. Steady shear rate experiments were conducted at 2000 s−1 (150 rpm) using the same stabilization parameters. The total assay time was 2230 s. The stiffness modulus was analyzed by the pseudo Wigner− Ville distribution (PWVD), which is the Fourier transform of the autocorrelation of the input signal. In this method, a signal s(t) is transformed to the time−frequency domain. The distribution may be considered a spectral density function in its three-dimensional (3D) form. The calculation algorithm was implemented in the Matematica 9 coding language. The Wigner distribution (WDF) for a continuous signal is expressed as −∞

w(t , ω) =

∫∞

s*(t − τ /2)s(t + τ /2)e−iωτ dτ

w(mΔt , k Δω) = RE{2Δt FFT[corr(i)]} where 1 ≤ i ≤ N + 1

⎧ s(m + i − 1)s*(m − i + 1) m ≥ i corr(i) = ⎨ ⎩ 0 m